Regression diagnostic

<p>In <a href="/facts/Statistics/DNPsKGYU">statistics</a>, a regression diagnostic is one of a set of procedures available for <a href="/facts/Regression_analysis/n6z5Tf7K">regression analysis</a> that seek to assess the validity of a model in any of a number of different ways. This assessment may be an exploration of the model's underlying <a href="/facts/Statistical_assumption/GWYDst34">statistical assumptions</a>, an examination of the structure of the model by considering formulations that have fewer, more or different <a href="/facts/Dependent_and_independent_variables/dINfzIF2">explanatory variables</a>, or a study of subgroups of observations, looking for those that are either poorly represented by the model (<a href="/facts/Outlier/VBhIzGbx">outliers</a>) or that have a relatively large effect on the regression model's predictions.
</p><p>A regression diagnostic may take the form of a graphical result, informal quantitative results or a formal <a href="/facts/Statistical_hypothesis_test/yv9RPF6U">statistical hypothesis test</a>, each of which provides guidance for further stages of a regression analysis.
</p>

Regression diagnostic open-in-new

Regression diagnostic